| WCCN聚类序列核函数在话者识别中的应用 |
| |
| 引用本文： | 邢玉娟,李恒杰,胡建军,王万军.WCCN聚类序列核函数在话者识别中的应用[J].云南大学学报(自然科学版),2013,35(2):167-172. |
| |
| 作者姓名： | 邢玉娟 李恒杰 胡建军 王万军 |
| |
| 作者单位： | 甘肃联合大学 电子信息工程学院 |
| |
| 基金项目： | 甘肃省教育厅基金项目(1113-01) |
| |
| 摘 要： | 针对说话人确认系统中GMM超向量建模计算复杂度高以及易受信道干扰的问题,提出一种新型的基于Bhattacharyya距离聚类的WCCN序列核函数算法.首先计算话者GMM模型之间的Bhattacharyya距离,根据该Bhattacharyya距离对话者模型进行聚类,得到聚类中心模型;紧接着对聚类中心模型的均值向量进行MAP自适应,进而生成超向量序列核函数;最后采用WCCN平滑归一化技术对序列核函数进行信道补偿,抑制噪音和信道畸变对核函数的影响.将该Bhattacharyya聚类WCCN核函数应用到SVM说话人确认系统,仿真实验结果表明该核函数可以有效地提高系统的识别准确率和识别速度.
|
| 关 键 词： | 语音识别 GMM超向量 Bhattacharyya距离 类内协方差归一化 支持向量机 |
A novel WCCN clustering kernel applied in speaker recognition |
| |
| XING Yujuan,LI Hengjie,HU Jianjun,WANG Wanjun.A novel WCCN clustering kernel applied in speaker recognition[J].Journal of Yunnan University(Natural Sciences),2013,35(2):167-172. |
| |
| Authors: | XING Yujuan LI Hengjie HU Jianjun WANG Wanjun |
| |
| Institution: | School of Electronics and Information Engineering,Gansu Lianhe University |
| |
| Abstract: | A novel WCCN kernel based on Bhattacharyya distance clustering algorithm was proposed in this paper in order to reduce the computation complexity of GMM super-vector,meanwhile the channel interference was removed from speaker verification system.Firstly,the GMM models of speakers were clustered based on Bhattacharyya distance,and clustering center models were obtained.Then super-vector sequence kernel was generated by adapting only mean vectors of these clustering center models.Finally,WCCN was used to restrain the noise and channel distortion effection of this kernel.Our experiment results showed that our new kernel can improve the recognition accuracy and speed. |
| |
| Keywords: | speech recognition GMM super-vector Bhattacharyya distance within class covariance normalization support vector machine |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《云南大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《云南大学学报(自然科学版)》下载免费的PDF全文 |
